Table 2 provides the descriptive statistics of the HII-SOP-K scores obtained by the SOs. Table 2 also provides the maximum possible scores that can be obtained in each subdomain and the total score. The “mean/total” row shows the percentage of third-party disability for each subscale and the total scale. Inter-item correlations of questions within each subscale and the total scale are provided in the last row of Table 2. A majority of correlations were weak to moderate [33]. The distribution of HII-SOP-K scores across different severity levels of third-party disability in SOs revealed that 21 individuals (36.8%) experienced mild third-party disability, 19 (33.3%) had no third-party disability, 13 (22.8%) had moderate disability, and 4 (7.0%) experienced severe third-party disability.

The translated scale demonstrated excellent internal consistency, with an overall Cronbach’s alpha of 0.90. Additionally, Cronbach’s alpha was calculated after each item was removed from the questionnaire. The values for Cronbach’s alpha, with a 95% confidence interval, ranged from 0.836–0.941, indicating that every item contributed significantly to the overall scale. A detailed breakdown of Cronbach’s alpha for each questionnaire item is provided in Supplementary Table 1 (in the online-only Data Supplement). An internal consistency assessment was also performed for each of the five factors that were obtained during factor analysis (described below in the next section). The Cronbach’s alpha for this analysis is reported in Table 3. Table 3 shows that the first three factors had good reliability, whereas last two factors had questionable reliability.

To validate and identify the underlying structure of the translated questionnaire, exploratory factor analysis (EFA) was performed via the principal axis factoring method with orthogonal varimax rotation. The Kaiser–Meyer–Olkin (KMO) test and Bartlett’s test of sphericity were used to assess the sampling adequacy. The KMO values should be greater than 0.5 [34], and the overall KMO value was 0.786 for the present data, confirming the adequacy of the sample for analysis. The significant results of Bartlett’s test of sphericity (χ²(190)=591.691, p<0.001) indicated that the correlations between items are sufficiently large for factor analysis [35]. Goodness of fit of the model was evaluated via the chi-square test. The chi-square test value was not statistically significant (χ²(100)=111.002, p=0.212), indicating a satisfactory fit of the obtained data to the proposed model. An additional root mean square error of approximation (RMSEA) model fit index value of 0.04 was obtained, which further strengthened the interpretation of good fit. The RMSEA is an absolute fit index that assesses how far a hypothesized model is from a perfect model. Researchers have suggested that the RMSEA values of less than 0.05 indicate that a model has close fit, while values in the range of 0.05 to 0.08 indicate fair fit and values greater than 0.10 indicate poor fit [36]. A cut-off value close to 0.06 is suggested by others [37].

In the present study, five factors were extracted based on the eigenvalue ≥1. Factor loadings greater than 0.4 were considered for each item to load onto a particular factor. Factor 1 comprised questions 7, 20, 4, 9, 13, 17, 5, and 14, which describe frustration, relationship dissatisfaction, and intimacy-related challenges. Factor 2 included questions 5, 14, 3, 11, and 16, which describe the emotions that can arise in SOs because of the individual’s hearing loss. Factor 3 consisted of 8, 18, and 12 questions, which describe social and lifestyle impacts and difficulties during social interaction. Factor 4 consisted of questions 6 and 19, which describe the additional effort required for communication, including visual cues and attentional strategies. Factor 5 consisted of questions 2, 15, and 1, which describe interactional adjustments, such as speaking louder or repetition or modifying environmental settings. Question 10 did not align with any of the identified factors. Additionally, questions 5 and 14 exhibited cross-loading (factor 1 and factor 2) (Table 4). The cumulative variance accounted for by these five factors was 58.3%. The variance accounted for by each factor is shown in Table 3.

Retests were conducted on a small sample of six individuals between seven days and one month after the initial assessment. The results of the ICCs for the three subscales and the total scores of the HII-SOP-K are shown in Table 5. ICC values ranging between 0.75 and 0.90 indicate good test–retest reliability, and values greater than 0.90 indicate excellent reliability [38]. Among the three subscales, relationship and emotions and social impact have excellent reliability, whereas communication strategies indicate good reliability. The total score suggests excellent test–retest reliability of the HII-SOP-K.

The mean difference between the test and retest HII-SOP-K scores was 7.08 (SD=4.31). The SD of the difference scores was multiplied by the critical value of an assumed normal distribution to calculate critical difference scores [14]. To obtain a 95% critical difference, the SD was multiplied by 1.96, which is equal to 8.44. Thus, the HII-SOP-K scores must change by at least 9 points for a true change to have occurred with 95% probability.

Understanding third-party disability across various relationships between SOs and IHL was of interest, although it was not the main objective of the study. This is important because the present study participants were not only spouses of IHL but also children and others. For this analysis, three groups of SOs were made—spouses, children, and others. Others included all other relationships apart from spouses and children. Fig. 1 shows the mean HII-SOP-K scores obtained for these three groups. The Kruskal–Wallis test revealed no significant difference in HII-SOP-K scores among these three groups (H(2)=0.626, p=0.731), indicating that the SOs in the present study experienced similar levels of third-party disability irrespective of their relationship with the IHL. We further probed into the subscale scores across the relationship categories with Kruskal–Wallis test, which suggested that the scores obtained by spouses, children, and others were not significantly different across the three subscales: relationship and emotions (H(2)=0.614, p=0.736), social impact (H(2)=0.502, p=0.778), and communication strategies (H(2)=0.017, p=0.992). Fig. 2 shows the raincloud plot of the scores obtained by spouses, children, and others across the three subscales. A raincloud plot provides individual participant data, summary statistics in the form of box plot, and the distribution of raw data. From Fig. 2, it can be noted that the median scores for the relationship and emotions were similar across spouses, children, and others. However, the IQR was large for others, followed by spouses and children. With respect to communication strategies subscale, the median and IQR for children and others were similar, whereas the median and IQR for spouses were slightly lower. Further, the median for social impact subscale was 0 for children and spouses, while others and spouses exhibited larger IQR.

The concept of third-party disability has been well documented, emphasizing that the effects of hearing loss extend beyond the individual to those closest to them. This study supports previous literature, indicating that family members of IHL report decreased QoL due to communication strain, social isolation, and changes in relationship dynamics [11,12]. The mean total score of 30.22 in the present study is very similar to that of 32.87 obtained in the original English version developed by Preminger and Meeks [14] and of 31.4 obtained by Rajasekaran, et al. [18] for the Indian population. Notably, the participants in both studies were only spouses, whereas the present study population included SOs other than spouses. In communication strategy domain of the present study, the mean score (13.68) is like that of English version (11.62). However, the Indian study reported a lower score (9.7) than the above two studies. When the mean social impact domain score of the present study (3.07) is compared to that obtained in the original English version (2.99) and that reported in the Indian population (6), it can be noted that the Indian study obtained higher score than other two studies. On the other hand, the mean score obtained for the relationship and emotions domain (13.46) in the present study was similar to that obtained by the Indian population (15.8), but lower than the score in the English version (18.26).

From the above comparisons, it can be observed that there are similarities and differences in the scores obtained among three studies. While the total scores are similar in these investigations, communication strategies and social impact subscale scores were similar in the present study and the English version. However, the relationship and emotions subscale scores were similar in the present and Indian studies and were lower as compared to English version. Though this cannot be statistically compared, the relationship and emotions domain appeared to be less affected in the present study sample. Hallam, et al. [39] reported that many family members make adjustments and necessary changes for the IHL with unquestioned commitment, respect, and consideration. However, this will not be the case in all families and relationships [40]. Whether the adjustments made by the SOs are perceived in a positive or negative way may depend on the prior state of the relationship and the personalities involved [39]. Considering the diversity of the relationship of SOs with IHL in the present study and small sample size, studies should be conducted on a large sample to validate the findings of present study. The present study aimed to validate the HII-SOP-K and did not attempt to investigate the nature and severity of third-party disability experienced by different SOs in Indian context.

The HII-SOP-K demonstrated high reliability and internal consistency (Cronbach’s alpha=0.90), which is similar to the original English version (Cronbach’s alpha=0.93). Additionally, Cronbach’s alpha was calculated after each item was removed from the questionnaire, which indicated that every item contributed significantly to the overall scale. Thus, the HII-SOP-K measures a single construct. These findings are consistent with the English version of the HII-SOP [14]. Hence, the HII-SOP-K can be utilized for the assessment of third-party disability among the SOs of IHL who speak and understand the Kannada language.

The original English version [14] had a total of three factors that explained 50.7% of the variance in the data. In contrast, the present HII-SOP-K has five factors that explain 58.3% of the total variance in the responses. This means that these five factors captured most of the differences in how the SOs responded to the questionnaire. Additionally, the percentage of variance explained is greater than that in the original version. The questions under each factor were noted to understand the underlying dimensions in HII-SOP-K. Notably, the first factor consisted of eight questions, seven of which belong to the category of “relationships and emotions” of the English version. In the second factor, all five questions belonged to the category of “relationships and emotions” of the English version. Notably, questions 5 and 14 demonstrated cross-loadings on factors 1 and 2, both of which were related to the “relationships and emotions” domain. Overall, factors 1 and 2 of HII-SOP-K represent the “relationships and emotions” dimension of third-party disability.

Factor 3 consisted of three questions, all of which were from the dimension of “social impact” of the English version. The last two factors of 4 and 5 consisted of three and two questions, respectively. These five questions explained the underlying dimension of “communication strategies.” The above comparison reveals that although the number of factors differed between the English version and the Kannada version, the underlying dimensions remain similar. These findings strengthen the validity of HII-SOP-K. Moreover, it is not uncommon to obtain different numbers of factors between different language versions of the same questionnaire. The reason could be due to linguistic and semantic differences, cultural and contextual adaptations, methodological variations, and translator interpretations. Additionally, the original text may have used broad categories that needed to be further defined for clarity or may have used different analytical approaches that resulted in different classifications [41,42].

An internal consistency analysis of the HII-SOP-K with questions retained as it is in the English version, revealed acceptable reliability for the “communication strategies” scale and good reliability for the other two scales. In total, the HII-SOP-K scale and its subscales have adequate internal consistency reliability, suggesting that the 20 items measure a single construct and that the subscales measure distinct subconstructs. In addition, consistent with the original scale, most of the inter-question correlations were less than 0.5, suggesting the unique contribution of each question to the total scale.

Both versions revealed very high test–retest reliability (0.90). Hence, similar to the HII-SOP, the Kannada version can be used in clinical settings to measure the effects of aural rehabilitation. A change of 9 points is considered significant for HII-SOP-K. This change is very small compared with the expected difference of 20 points observed in HII-SOP. A probable reason for this difference could be the small sample size used for the retests in the present study. Hence, the present critical value can be considered a preliminary finding, and the replicability of test–retest results in a large sample should be done to confirm if this critical value is consistent.

Since the English version of the HII-SOP was validated in spouses aged between 23 and 82 years, Preminger and Meeks [14] proposed that the HII-SOP is appropriate for spouses across all ages. The age range of the present study participants was similar to that of the original study (20–85 years). Thus, HII-SOP-K can be considered appropriate for measuring third-party disability in SOs across all ages. Furthermore, the present study participants had diverse educational backgrounds. Hence, HII-SOP-K is applicable to those SOs who are proficient in reading and understanding Kannada. In the present study population, the SOs included not only spouses of the IHL but also their children and others. Thus, HII-SOP-K is suitable for any SO who is living with the IHL and is interacting with them daily.

The literature shows that spouses of IHLs report frustration due to frequent communication breakdowns; irritation and embarrassment related to the impact of hearing loss in social situations; and tension, worry, and concern about the problems experienced by their partner with hearing loss [7,12,43]. On the other hand, Preminger, et al. [44] explored the third-party disability experienced by adult children with IHL. The findings revealed that adult children often face significant challenges in communication, requiring increased effort that leads to negative emotional outcomes, such as irritability and frustration. The study also indicated that adult children experienced a greater sense of uncertainty regarding their parents’ health and communication needs than their spouses did. The authors attributed this discrepancy to the fact that adult children typically do not live together with their parents or accompany them to medical appointments, which limits their access to information and support. In the Indian context, many adult children live with their parents, leading to different relationship dynamics between parents and children.

Manchaiah, et al. [40] reported that communication partners such as friends, colleagues, and caregivers also experienced significant emotions such as frustration, anger, and embarrassment, along with a notable reduction in social interactions stemming from communication breakdowns. Furthermore, they emphasized that the spouses of IHL were most affected across the communication, social, and emotional dimensions. However, in the present study, a comparison of three subscales of the HII-SOP-K among the three categories of SOs did not reveal any significant differences. Manchaiah, et al. [40] hypothesized that SOs who appeared to have a positive attitude towards life reported fewer psychosocial consequences and coped well. Such SOs also had some experience in dealing with other chronic conditions. Further, research also suggests that there is considerable variation in how caregivers adapt to their care-giving demands [45]. The experiences of SOs change over time as they adapt to the hearing loss of their family member [40]. Thus, several factors could have influenced the HII-SOP-K score of the present study sample. Studies are warranted on a large sample across diverse types of SOs to understand the third-party disability experienced at different phases in the journey of SOs as communication partners of IHL.

SOs typically understand the IHL and experience the communication difficulties caused by the hearing impairment on a daily basis [46]. Since any SO of IHL can experience third-party disability, a tool assessing this construct should be applicable to all SOs who might communicate with IHL. During the process of translation adaptation, care is taken in the choice of words such that all the questions are equally applicable to any SO of the individual with hearing loss and not specifically applicable to spouses only. Taken together, the applicability of HII-SOP-K extends beyond the scope of spouses and is suitable for assessing third-party disability in any SO of an IHL.

The experiences of SOs of IHL change over time. Understanding the different phases of adjusting to the hearing loss of their family members by SOs is important for audiologists during counseling, such that the information provided is tailored to meet the individual SO’s needs [40]. Exploring the effects of different variables such as age, gender, the prior quality of the relationship, personality characteristics, and coping with other medical conditions or sensory impairments is of questions of research and clinical interest. Thus, HII-SOP-K can be used in both research and clinical settings to understand the third-party disability experienced by SOs of IHL who speak the Kannada language.

The current study translated and adapted the HII-SOP into the Kannada language. The SOs were spouses, children, siblings, or other significant family members of IHL. Overall, all the measures indicated that the translated HII-SOP-K possesses desirable psychometric properties and is a reliable and valid tool for assessing third-party disability in the SOs of IHL who speak the Kannada language.